3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
checkpoint52a_post |
7 |
|
o modifications to make FREEZE (_SURFACE) flux visible to pkg/kpp |
8 |
|
- moved surfaceTendencyTice from pkg/seaice to main code |
9 |
|
- subroutine FREEZE_SURFACE now limits only surface temperature |
10 |
|
and is called in FORWARD_STEP (if allowFreezing=.TRUE.) |
11 |
|
=> new output.txt for global_ocean.cs32x15 & global_with_exf |
12 |
|
- keep the old version of FREEZE for backward compatibility; |
13 |
|
global_ocean.90x40x15 left unchanged using useOldFreezing=.TRUE. |
14 |
|
- corresponding modifs for TAF directives |
15 |
|
o added surface flux output variables to TIMEAVE_STATVARS |
16 |
|
o time-averaged output for pkg/ptracers |
17 |
|
o fix problem with ECCO-related divided adjoint |
18 |
|
snapshot file was missing some ad-variables, |
19 |
|
but no warning issued by TAF |
20 |
|
|
21 |
|
checkpoint52a_pre |
22 |
|
o some bug fixes for #undef REAL4_IS_SLOW |
23 |
|
|
24 |
|
checkpoint52 = ecco_c52_e35 |
25 |
|
o after testing adjoints of checkpoint51u_post |
26 |
|
|
27 |
|
checkpoint51u_post |
28 |
|
o this is the post-merge tag of the ecco-branch |
29 |
|
(needs another clean checkout and re-testing) |
30 |
|
o various natl_box_adjoint configs and results |
31 |
|
but no input fields |
32 |
|
o obcs adjoint not yet merged! |
33 |
|
o adding ECCO global 1x1 and 2x2 degree configs |
34 |
|
o updated/extended some store directives for seaice |
35 |
|
o pkg/ecco now containes ecco-specific part of cost function |
36 |
|
o top level routines the_main_loop, forward_step |
37 |
|
supersede those in model/src/ |
38 |
|
previous input data.cost now in data.ecco |
39 |
|
(new namelist ecco_cost_nml) |
40 |
|
o in ctrl/ cleaned some cross-dependencies and updated CPP options |
41 |
|
|
42 |
|
checkpoint51t_post |
43 |
|
o this is the pre-merge tag of the ecco-branch |
44 |
|
|
45 |
|
checkpoint51s_post |
46 |
|
o cleanup: removed various obsolete #define-s |
47 |
|
o pkg/grdchk |
48 |
|
- renamed 'ph-grd 3' to 'grad-res' |
49 |
|
- merged from ecco-branch |
50 |
|
(remaining bug fixes for obcs gradient checks) |
51 |
|
- additional high-precision output for testreport |
52 |
|
(grep for 'precision_grdchk_result') |
53 |
|
|
54 |
|
checkpoint51r_post |
55 |
|
o cleaned up or removed CPP_OPTIONS.h files |
56 |
|
o updated results for some of the tests to reflect the new |
57 |
|
optfiles (-ieee: -O0 instead of -O3) |
58 |
|
|
59 |
|
checkpoint51q_post |
60 |
|
o cd_code: |
61 |
|
- moved cd_scheme.F -> cd_code_scheme.F |
62 |
|
- separate read_checkpoint from cd_code_ini_vars.F |
63 |
|
- separated cd_code part from write_checkpoint |
64 |
|
- updated AD_SOURCE, generated .flow |
65 |
|
- added CD_CODE_VARS.h to the_main_loop |
66 |
|
- Moving cd_code specific store dir.s to pkg/cd_code |
67 |
|
o ctrl package totally restructured |
68 |
|
- pack/unpack now optional and decoupled from |
69 |
|
xx_/adxx_ I/O |
70 |
|
- ctrl_pack/unpack cleaned |
71 |
|
(new routines ctrl_init_ctrlvar.F, pkg/ctrl/ctrl_init_wet.F) |
72 |
|
- confined inclusion of AD_CONFIG.h to where necessary. |
73 |
|
o genmake2: changed AD dependencies from SRCFILES to AD_FILES |
74 |
|
|
75 |
|
checkpoint51p_post |
76 |
o create initial cd_code package |
o create initial cd_code package |
77 |
|
- WARNING: the initialization of fields within packages |
78 |
|
is broken because INI_FIELDS (where READ_CHECKPOINT is |
79 |
|
called) is called *before* PACKAGES_INIT_VARIABLES. |
80 |
|
This seems to be backwards since |
81 |
|
1) its important to initially zero the entire field including |
82 |
|
the "halo" region |
83 |
|
2) zeroing needs to be done *before* reading pickup info |
84 |
|
and the pickup call might or might not happen in any |
85 |
|
particular run |
86 |
|
- this needs an overhaul for R2 |
87 |
o cleanup verification/*/code/GMREDI_OPTIONS.h |
o cleanup verification/*/code/GMREDI_OPTIONS.h |
88 |
o cleanup testreport |
o cleanup testreport |
89 |
|
|